Expression Atlas - a Gene and Protein Expression Look-up Tool

Biological science has seen a steady growth of "big data" in transcriptomics and proteomics over the past decade. With that is a growing need for a user-friendly platform to look up expression data via gene/protein of interest, sample properties (e.g. cell line, species name, disease name) or main intent of an experiment (e.g. comparing knockouts vs wild-types, treated vs untreated samples), especially for researchers who do not have access to high-performance computing facilities to analyse very large data sets. Expression Atlas at EMBL-EBI fills this role by providing pre-analyzed RNA-seq, expression microarray and proteomics data for researchers to discover and query which genes/proteins are expressed in which tissues, cell types, developmental stages, and how expression changes under many other experimental conditions.

In this talk, I will show examples of querying the Atlas in a baseline context, e.g. find genes expressed in the bovine kidney, or in a differential context, e.g. find genes that are up/downregulated in response to salt stress in rice. I'll also illustrate how the combination of manual curation of meta-data, a standardised statistical pipeline and the consumption of Ensembl gene annotation data improves reproducibility of the analysed results.

As of December 2016, the Atlas consists of 3023 datasets, 485 of which are RNA-seq. All data in Expression Atlas are free for download/reuse, and are sourced from two archival databases at EMBL-EBI: ArrayExpress for transcriptomic data via a user-friendly webform-based submission tool Annotare, and PRIDE for proteomics data, the latter a new addition to Atlas in 2016.